Acoustic Processing Method and Acoustic Device

1. An acoustic processing method comprising: extracting a voice feature amount from a first audio signal representing sound collected by a microphone; receiving a second audio signal from a device at a far-end side; calculating a noise feature amount based on a power level of the first audio signal, in a state where: the power level of the first audio signal is less than or equal to a first predetermined value; and a power level of the second audio signal is less than a second predetermined value, determining a gain of the second audio signal based on the extracted voice feature amount and the calculated noise feature amount; and adjusting the second audio signal based on the determined gain of the second audio signal.

2. The acoustic processing method according to claim 1 , further comprising: removing an echo element from the first audio signal using an echo canceller, wherein the extracting extracts the voice feature amount from the first audio signal after the removing has removed the echo element from the first audio signal.

3. The acoustic processing method according to claim 1 , further comprising: removing an echo element from the first audio signal using an echo canceller, wherein the calculating calculates the noise feature amount from the first audio signal after the removing has removed the echo element from the first audio signal.

4. The acoustic processing method according to claim 1 , wherein the voice feature amount includes the power level of the first audio signal.

5. The acoustic processing method according to claim 1 , wherein the determining distinguishes a voice section from the first audio signal and determines the gain of the second audio signal from the distinguished voice section.

6. The acoustic processing method according to claim 1 , further comprising outputting the adjusted second audio signal to a speaker.

7. An acoustic processing method comprising: obtaining a noise level from a first audio signal representing sound collected by a microphone; receiving a volume adjustment from a user; storing the obtained noise level as a previous noise level, in a state where a noise section has been distinguished and the receiving has received the volume adjustment from the user; receiving a second audio signal from a device at a far-end side; determining a gain of the second audio signal based on the stored previous noise level and the obtained noise level that is currently obtained, in a state where the volume adjustment from the user is not received; and adjusting the second audio signal based on the determined gain of the second audio signal, wherein the noise section is distinguished in a case where: a power level of the first audio signal is less than or equal to a first predetermined value; and a power level of the second audio signal is less than a second predetermined value.

8. The acoustic processing method according to claim 7 , further comprising: removing an echo element from the first audio signal, wherein the obtaining obtains the noise level from the first audio signal after the removing has removed the echo element from the first audio signal.

9. An acoustic device comprising: a microphone; at least one memory storing instructions; a processor that implements the instructions and executes a plurality of tasks including: a noise level obtaining task that obtains a noise level from a first audio signal representing sound collected by the microphone; a volume adjusting task that receives a volume adjustment from a user; a storing task that stores the obtained noise level as a previous noise level in the at least one memory, in a state where a noise section has been distinguished and the volume adjustment task receives the volume adjustment from the user; an audio signal receiving task that receives a second audio signal from a different device at a far-end side; a gain determining task that determines a gain of the second audio signal based on the stored previous noise level and the obtained noise level that is currently obtained, in a state where the volume adjustment task has not received the volume adjustment from the user; and a level adjusting task that adjusts the second audio signal based on the determined gain of the second audio signal, wherein the noise section is distinguished in a case where: a power level of the first audio signal is less than or equal to a first predetermined value; and a power level of the second audio signal is less than a second predetermined value.

10. The acoustic device according to claim 9 , wherein: the plurality of tasks include an echo cancelling task that removes an echo element from the first audio signal, and the noise level obtaining task obtains the noise level from the first audio signal after the echo cancelling task has removed the echo element from the first audio signal.

11. The acoustic device according to claim 9 , wherein the gain determining task detects a voice section from the first audio signal and determines the gain of the second audio signal from the detected voice section.

12. The acoustic device according to claim 9 , wherein the gain adjusting task outputs the adjusted second audio signal to a speaker.